The effect of heat conduction on the realization of the primary standard for sound pressure

Abstract

Pressure reciprocity calibration of microphones provides the basis for primary measurement standards for sound pressure in air. At low frequencies, reciprocity calibration requires that a heat conduction correction be employed to account for energy transfer to and from the bounding surfaces of the close-coupled microphone arrangement. The standard governing reciprocity calibration, IEC 61094-2 : 2009, provides two models for the heat conduction correction: the Low Frequency Solution, and the Broadband Solution. Analysis has revealed significant and unexplained differences in behaviour between the models at very low frequencies, leading to inconsistency in calibration results, which has been quantified. Additionally, both heat conduction solutions given in IEC 61094-2 are simplifications that strictly apply only above their respective lower limiting frequencies. An international comparison on microphone calibration is currently underway that includes measurements below the lower limiting frequencies of the models. In this paper, the origin and nature of the Broadband simplifications have been identified, and estimates of the error given. A flaw in the Broadband theory is identified and its effect quantified. Simplification error for the Low Frequency solution is evaluated, and the full spectrum solution is given. This paper urges caution in the application of the models at low frequency and provides data useful for assessing the contribution to the measurement uncertainty.